liforma docs

Text-to-Speech

Natural speech synthesis from text.

What is it?

Convert text to natural speech with Liforma's voice pipeline. No avatar rendering, no conversation loop — audio only.

When to use it

  • Voice prompts, IVR, or accessibility narration
  • Apps that already have visuals and only need speech
  • Batch or server-side audio generation
  • When lip-sync and character animation are not required

API

import { Liforma } from '@liforma/client';

const audio = await Liforma.textToSpeech({
  text: 'Your table is ready.',
  voiceId: 'en-GB-SoniaNeural'
});

// audio.buffer — playable audio
// audio.visemes — optional timing data

When to use Avatar Experiences instead

Use an Avatar Experience when speech should come from a visible character in a live, interactive session. TTS is included in every experience — this standalone API is for audio-only integrations.